The present invention relates to a disk drive and a method of controlling a disk drive, and more particularly to a head positioning mechanism for a disk drive and a method of controlling the disk drive.
Disk drives, particularly magnetic disk drives, are widely used by computers to access disks as external information storage medium providing a large storage capacity.
Generally, a magnetic disk drive system has as its main components, a plurality of storage medium disks concentrically and rotatably supported, a plurality of heads facing each disk surface, a head positioning mechanism for supporting each head and moving it to a target track, and a controller for controlling the operation of the head positioning mechanism.
In most of the methods applied to positioning a head, the head position is controlled by using a velocity control and a position control alternately. Specifically, first a head is moved near to a target track at a high velocity under control of a speed feedback control system which gives a velocity command corresponding to the number of tracks from the present head position to a target track and makes the head move at given speed. Then the velocity feedback control system is switched to a position feedback control system by which the head is finely positioned at the center of the target track.
In a related apparatus disclosed in JP-A-2-56008, there is proposed a magnetic disk drive having a position control system which includes a head positioning mechanism (a controlled object), a compensator, and a head position detector, wherein a filter is serially connected to the position control system at the front stage thereof, the filter having an inverse characteristic of the control system dynamics in order to obtain a desired output response having a transfer function of one for a signal path from the filter input to the control system output.
The following problems have been found in the above-described position control systems for magnetic disks.
(1) In most of the conventional methods whereby speed control is switched to position control, a single type of speed command is used over an operation range from a short to a long head seek distance. In such a case, particularly in the case of a short head seek distance, it is difficult for an actuator to generate a large output, and so to move the head at a sufficiently high speed.
(2) In the conventional method of switching between the control systems, acceleration of the controlled object changes rapidly at the switching point, so that a head and head positioning mechanism are mechanically vibrated, thereby requiring a long head position alignment time.
(3) In order to provide a filter having an inverse characteristic of a head positioning system, it becomes necessary to prepare a model of a controlled object. However, the head positioning mechanism to be controlled has in many cases complicated vibration modes, so that a precise model cannot be obtained. Thus, the transfer function is not one at the signal path from the input of the filter having the inverse dynamic characteristic to the output of the control system. If an input is designed without considering such a transfer function, the resonance point of a head positioning mechanism is excited, resulting in a long head seek time.